Koplin Julian J, Johnston Molly, Webb Amy N S, Whittaker Andrea, Mills Catherine
Monash Bioethics Centre, Monash University, Clayton, VIC, Australia.
School of Social Sciences, Monash University, Clayton, VIC, Australia.
Hum Reprod. 2025 Feb 1;40(2):179-185. doi: 10.1093/humrep/deae264.
Artificial intelligence (AI) has the potential to standardize and automate important aspects of fertility treatment, improving clinical outcomes. One promising application of AI in the fertility clinic is the use of machine learning (ML) tools to assess embryos for transfer. The successful clinical implementation of these tools in ways that do not erode consumer trust requires an awareness of the ethical issues that these technologies raise, and the development of strategies to manage any ethical concerns. However, to date, there has been little published literature on the ethics of using ML in embryo assessment. This mini-review contributes to this nascent area of discussion by surveying the key ethical concerns raised by ML technologies in healthcare and medicine more generally, and identifying which are germane to the use of ML in the assessment of embryos. We report concerns about the 'dehumanization' of human reproduction, algorithmic bias, responsibility, transparency and explainability, deskilling, and justice.
人工智能（AI）有潜力使生育治疗的重要方面标准化和自动化，从而改善临床结果。人工智能在生育诊所的一个有前景的应用是使用机器学习（ML）工具来评估用于移植的胚胎。要以不损害消费者信任的方式成功在临床上应用这些工具，就需要意识到这些技术引发的伦理问题，并制定策略来处理任何伦理问题。然而，迄今为止，关于在胚胎评估中使用机器学习的伦理问题的已发表文献很少。这篇小型综述通过更广泛地审视机器学习技术在医疗保健和医学中引发的关键伦理问题，并确定哪些问题与在胚胎评估中使用机器学习相关，为这个新兴的讨论领域做出了贡献。我们报告了对人类生殖“非人性化”、算法偏差、责任、透明度和可解释性、技能退化以及公平性等问题的担忧。
